Abstract: How to improve the acoustic characteristics of chord type instrument resonators has been a hot issue in the industry of instrument making. Taking the guitar as an example, this paper presented a verification method of vibration characteristics of guitar based on EMI technology.  A set of finite element models of the resonant box were built. By adjusting its boundary parameters, the main boundary parameters which affect the vibration characteristics were determined by modal analysis. The experimental platform of vibration characteristic impedance and acoustic analysis of the resonance box was built, and the vibration characteristic impedance and acoustic experiments of the resonance box were completed. It is concluded that EMI technology can be used to improve the vibration characteristics of guitar . Finally, the diameter of string was predicted by neural network，it was concluded that the neural network can be applied to the prediction of guitar string diameter.

Key words: piezoelectric impedance technology, guitar, vibration characteristics, modal analysis, natural frequency, acoustic experiment, the neural network

摘要： 如何改善弦类乐器共鸣箱的声学特性一直是乐器制作行业亟待解决的热点问题。以吉他为例，提出了一种基于压电阻抗技术的吉他振动特性的验证方法，搭建了一组吉他的有限元模型，借助于调整其边界参数进行模态分析，确定了影响其振动特性的主要边界参数；搭建了吉他的振动特性阻抗及声学分析实验平台，完成吉他振动特性的阻抗及声学实验，得到了压电阻抗技术可以用于改善吉他共鸣箱的振动特性的结论；最后应用神经网络对弦的直径进行预测，得出神经网络可以运用到吉他弦直径的预测。

关键词: 压电阻抗技术, 吉他, 振动特性, 模态分析, 固有频率, 声学实验, 神经网络